bestmodel.run / model — Qwen3-8B (agent view)

category chat · Qwen/Qwen3-8B · 44 runs in the model record
answer: 52.3 tok/s (model median)

Honesty ladder: measured > reported > extrapolated > formula > no data yet.
Empty fields stay empty — never filled with an estimate.

  rig | quant | median result | basis | ttft | peak vram | max context
  AMD Ryzen 7 7840HS                 | 4-bit | 12.3 tok/s     | reported | ttft -ms | vram - GB | ctx 512
  GTX 1080 Ti 11GB                   | 4-bit | 50.7 tok/s     | reported | ttft -ms | vram 7.8 GB | ctx 8,192
  GTX 1080 Ti 11GB                   | 5-bit | 36.8 tok/s     | reported | ttft -ms | vram 8.5 GB | ctx 8,192
  Multi-GPU 24GB ×2                  | 4-bit | 113.5 tok/s    | reported | ttft 186ms | vram - GB | ctx 2,048
  Radeon 890M 32GB                   | 4-bit | 15.9 tok/s     | reported | ttft -ms | vram - GB | ctx 4,096
  RTX 3060 12GB                      | 4-bit | 43.7 tok/s     | measured | ttft 370ms | vram - GB | ctx 131,072
  RTX 3060 Ti 8GB                    | 4-bit | 53 tok/s       | measured | ttft 1,206ms | vram 7.5 GB | ctx 65,536
  RTX 3090 Ti 24GB                   | 4-bit | 133 tok/s      | reported | ttft 123ms | vram 7 GB | ctx 8,192
  RTX 5090 32GB                      | 4-bit | 227.1 tok/s    | reported | ttft 18ms | vram 5.6 GB | ctx 8,192